Inflammatory changes of the endometrium in patients with minimal-to-moderate endometriosis*t

Transcription

1 FERTILITY AND STERILITY Vol. 62, No.5, November 1994 Copyright e 1994 The American Fertility Society Printed on acid.. free paper in U. S. A Inflammatory changes of the endometrium in patients with minimal-to-moderate endometriosis*t Maria C. Leiva, M.D.:j: Lisa A. Hasty, M.D. C. Richard Lyttle, Ph.D. II Department of Obstetrics and Gynecology, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania Objective: To investigate the endometrium of normal patients for chemotactic activity to neutrophils and macrophages and compare these findings with those of patients with endometriosis. Study Design: Endometrial biopsies from patients with and without endometriosis were analyzed for chemotactic activity throughout the menstrual cycle. Glands and stroma of luteal samples were isolated to determine the source of this activity. Infiltrating cells to the endometrium were identified by immunohistochemistry with the use of the monoclonal antibody OKMl. Results: Luteal endometrial samples from normal patients had higher chemotactic activity than samples from the proliferative phase; this was true for both cells types studied: macrophages, 84 versus 10 and neutrophils, 74 versus 11. Patients with endometriosis had high chemotactic activity in both proliferative and luteal biopsies: macrophages, 73 ± 9 versus 78 ± 1 and neutrophils, 41 ± 18 versus 63 ± 26. Stromal cells from luteal biopsies demonstrated a higher chemotactic activity than the epithelial component. Immunohistochemistry staining identified the infiltrating cells as macrophages. Conclusions: Extracts from endometrium of normal patients contain chemotactic activity for neutrophils and macrophages; this activity is higher in the secretory phase of the cycle. Endometriosis patients had high chemotactic activity throughout the menstrual cycle. Separation of the endometrial tissue into stroma and the glandular epithelium indicated that the stromal cells are the source of this factor. Fertil Steril1994;62: Key Words: Endometriosis, infertility, chemotaxis, macrophages, endometrium, menstrual cycle The observed inflammatory and immunologic changes in patients with endometriosis have been Received November 9, 1993; revised and accepted June 3, *Supported in part by grants HD20025 and HD06274 from the National Institutes of Health, Washington, D.C. t Presented at the 73rd meeting of The American Fertility Society, New Orleans, Louisiana, November 2 to 5, :j: Present ddress: Department of Obstetrics and Gynecology, Pennsylvania Hospital, Philadelphia, Pennsylvania. Present address: Department of Gynecology and Obstetrics, Division of Reproductive Endocrinology, Emory University, Atlanta, Georgia. II Reprint requests: C. Richard Lyttle, Ph.D., Women's Health Institute, Wyeth-Ayerst Research, 145 King of Prussia Road, Radnor, Pennsylvania (FAX: ). postulated to be responsible for the associated infertility of these patients, as well as for the pathogenesis of the disease. These alterations have been well documented and characterized in the peritoneal fluid (1-3). However only recently have investigators begun to study the eutopic endometrium of these patients. Previous studies from our laboratory (4) have demonstrated that glandular cells from endometriomas synthesize and secrete complement component C3. One significant observation of these studies on the expression of C3 is that the endometrium of patients with endometriosis displayed C3 synthesis in contrast to the endometrium of normal patients, which did not express C3 (5). More recent studies from our laboratory demonstrate that prolif- Vol. 62, No.5, November 1994 Leiva et al. lnfiammatory changes of the endometrium in endometriosis 967

2 erative endometrium from normal patients does not express C3 and several regulatory complement factors such as factor B, decay accelerating factor and complement receptor type 1 (CR1). However, when followed through the menstrual cycle, these proteins were expressed in the luteal phase (6). These results are in agreement with several other studies (7) that have demonstrated similar findings regarding the differences in the expression of endometrial proteins between normal endometrium and the endometrium of patients with endometriosis. Several endometrial markers also have been shown to be regulated hormonally. Endogenous peroxidase activity increases during the secretory phase of the cycle, which also correlates with an increase in the number of granulocytes present in the endometrial stroma. The majority of these endometrial granulocytes have been identified histochemically and are indistinguishable from neutrophilic leukocytes (8). This increase in granulocytes is observed also in the postpartum and peri-implantation periods, and yet its role has not been established clearly (9, 10). We have demonstrated recently (3) that patients with minimal-to-moderate endometriosis exhibit increased chemotactic activity of their PF toward macrophages and neutrophils; furthermore, this activity is suppressed after medical treatment of the disease. The demonstration of an increased chemotactic activity agrees with previous reports (11-13) of an increase in the number of macrophages and their activated products in the PF of these patients. The contribution of these cells to the pathophysiological modifications in the peritoneal cavity of these patients is yet to be determined, but the presence of greater number of cells can have a deleterious effect on the sperm-egg interaction or may be even responsible for phagocytosis of the sperm. Finally the inflammation observed could induce modifications in tubal function or the secretory products of these infiltrating cells could be toxic for the early embryo (14). In this study we looked at endometrium to investigate chemotactic activity in normal tissue to determine whether a pattern of hormonal regulation for this activity could be demonstrated. We examined also the eutopic endometrium of patients with minimal-to-moderate endometriosis to investigate any differences in chemotactic activity from that of normal patients. Finally, we identified the source of this activity as well as the nature of the majority of the infiltrating cells. MATERIALS AND METHODS These studies were reviewed by the institutional review board of the Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania, and the need for informed consent was waived. Endometrial biopsies were obtained from women during their infertility evaluation at the Hospital of the University of Pennsylvania, specimens were obtained from patients with a previous diagnosis of minimal-to-moderate endometriosis as defined by the classification of The American Fertility Society (15) as well as from disease-free patients. This group included patients with pure male factor infertility as well as patients with a documented ovulatory disorder who also had a normal pelvis. Luteal biopsies were obtained on cycles when no other diagnostic procedure was being performed. Patients undergoing diagnostic laparoscopy during the proliferative phase of the cycle were biopsied also. A total of 53 biopsies from 52 patients were analyzed. The distribution of the patients was as follow: 34 biopsies from endometriosis-free patients, of which 18 were obtained during the proliferative phase of the cycle and 16 in the secretory phase. Fourteen biopsies were obtained from patients with endometriosis, of which 10 were obtained in the proliferative phase and 4 in the secretory phase. For evaluation in the chemotactic assay, endometrial tissue was homogenized at 4 oc in 10 mm Tris-HCl (T10), ph 7.4, at a concentration of 50 mg/ml of buffer. Additional immunohistochemistry studies were performed in five samples from normal patients and four from patients with endometriosis. Chemotaxis was tested using a 48-well modified Boyden chamber with a 5-mm polycarbonate filter (Neuro-Probe, Cabin John, MD) as previously described (3, 16). Cells tested for chemotactic activity included the HL60 and U937 cell lines. The human leukemia cell line HL60-C15 has been shown previously (17) to differentiate to neutrophils by culture for 7 days in the presence of 1.27% dimethylsulphoxide (DMSO; Sigma Chemicals, St. Louis, MO). The human histiocytic leukemia cell line U937 differentiates to macrophages after stimulation with 1 mm dibutyril cyclic adenosine 5-monophosphate (Sigma Chemicals) for 72 hours (18, 19). After stimulation, cells were washed twice in phosphate-buffered saline (PBS) at room temperature, viability was assessed using 0.5% trypan-blue exclusion, and differentiation was checked by centrifugation on a cytospin model 1 (Shandon Southern Instruments, 968 Leiva et al. Inflammatory changes of the endometrium in endometriosis Fertility and Sterility

3 Sewickley, PA) for 10 minutes at 500 rpm and staining with Wright-Giemsa (EMI Diagnostics, Gibbstown, NJ). Purity and differentiation were usually ~90%. For the chemotaxis assay cells usually were used at a concentration of 1 X 10 6 cells/mi. Chemotaxis was assessed by counting under high power magnification the number of cells that migrated and attached to the lower side of the filter. Samples were run in triplicate, and all experiments were repeated at least three times. Endometrial glands and stroma were isolated from biopsy specimens using the procedure previously described by Satyaswaroop (20). Briefly, endometrial specimens were minced and incubated at 37 C with 0.25% collagenase type II (Sigma Chemicals) for 1 hour; after digestion, glands were separated from stromal cells by sieving and sedimentation. Purity of the cell preparation was assessed by microscopic examination and stromal cells were centrifuged in the presence of 30% percoll (Pharmacia, Piscataway, NJ) to remove contaminating blood cells. Cell preparations were resuspended and homogenized in T10, as described above, for use in the chemotaxis assays. Immunohistochemistry was performed with the use of the monoclonal antibody OKMI against complement receptor 3 (CR3). This antibody identifies neutrophils and macrophages in the endometrium. Briefly, fixed sections were deparaffinized and endogenous peroxidase was blocked by incubation in methanol and hydrogen peroxide solution. Blocking of nonspecific binding was accomplished with normal horse serum; then sections were incubated with the primary antibody and a biotynilated anti-mouse antibody (Vector Laboratories, Norwalk, CT) was used as second antibody. After washes with PBS, the avidin-biotin complex was added, slides were washed again, and Chromagen (0.05 DAB [p-dimethylaminoazobenzene]) was added; counterstaining was accomplished by incubation with hematoxylin. Final sections were mounted and examined under light microscope (6). All results are presented as means± SD. Statistical analysis was performed by means of Student's t-test and statistical significance was considered for P values s0.05. RESULTS The data presented in Figure 1 indicate that extracts from luteal phase endometrial biopsies from normal patients displayed the highest chemotactic activity for neutrophils differentiated from HL := 60.!!! (.) Prolif Luteal Figure 1 Chemotactic activity of extracts from normal proliferative and luteal endometrium. Chemotactic activity was determined using neutrophils differentiated from HL60 cells ( ) as well as macrophages differentiated from U937 cells (~). Results are expressed as mean number of cells (±SD). cells (74 ± 22) as well as macrophages derived from U937 cells (84 ± 10), whereas samples from the proliferative phase had significantly lower numbers of migrating cells: neutrophils, 11 ± 8; macrophages, 10 ± 2 (P < for both cell types). Patients with minimal-to-moderate endometriosis exhibited high chemotactic activity uniformly throughout the menstrual cycle: neutrophils in proliferative phase 41 ± 18 and in luteal phase 63 ± 26 (P = 0.07). The number of macrophages that migrated also were similar for both phases of the cycle: 73 ± 9 in the proliferative phase and 78 ± 1 for the luteal samples (P = 0.43) (Fig. 2). Figure 3A represents the immunohistochemical staining of endometrial sections of patients with minimal-to-moderate endometriosis; Figure 3B represents sections from disease-free patients. Both sections were stained with the monoclonal antibody to CR3, which also recognizes leukocytes. Sections from patients with endometriosis demonstrated the presence of increased numbers of infiltrating leukocytes in the stroma when compared with those of disease-free patients. The majority of the infiltrating cells were monocytic ( 40 ± 5 positive cells per section) and distributed throughout the endometrium. In contrast, endometrial sections from patients without the disease did not demonstrate a significant number of CR3-positive cells ( 4 to 5 per section). Further analysis of the chemotactic activity using isolated stromal and glandular epithelial cells Vol. 62, No.5, November 1994 Leiva et al. Inflammatory changes of the endometrium in endometriosis 969

4 from luteal specimens demonstrated a higher chemotactic activity for the stromal component with 98 ± 5 macrophages migrating versus 26 ± 8 for the glandular epithelium (P < 0:001) (Fig. 4). DISCUSSION Extracts from the endometrium of normal patients contain a chemotactic factor for macrophages. This activity is greater in the endometrium of patients in the luteal phase of the menstrual cycle compared with the proliferative phase. These results may suggest a possible role for steroid hormones in the regulation of this chemotactic factor. This possibility would be in agreement with the concept that steroid hormones are able to regulate cellular infiltration in the reproductive tract in other species. Previous results from our laboratory (16, 21) have demonstrated the estrogen regulation of an eosinophil chemotactic factor in the rat uterus. The separation of the endometrial tissue into stromal and glandular epithelium indicated that the stromal cells are the source of this chemotactic factor. Again, this is in good agreement with our findings in the rat uterus, where the stromal cells are the site of origin of the eosinophil chemotactic factor (Leiva MC, Xu Q, Fischkoff SA, Lyttle CR, abstract). Although these experiments suggest that the stromal cells may be the source of the factor, G) 3: ~ (J Pro li f Luteal Figure 2 Chemotactic activity of endometrial extracts obtained from patients with endometriosis. Extracts were prepared from both the proliferative and luteal phases of the menstrual cycle. Chemotactic activity was evaluated using neutrophils ( ) and macrophages ( ~) as in Fig. 1. Results are presented as mean number of cells (±SD). 970 Leiva et al. Figure 3 Immunohistochemical staining of proliferative endometrium biopsies obtained from patients with endometriosis (A) or disease-free patients (B). Immunohistochemistry was performed with the use of the monoclonal antibody OKM-1 (X40 magnification). further experiments are required to demonstrate that these cells are in fact responsible for the production of this chemoattractant. In contrast to the findings described for normal patients, patients with minimal-to-moderate endometriosis displayed a high chemotactic activity during both the proliferative and luteal phases of the cycle. Thus, no pattern of hormonally regulated chemotactic activity was demonstrated in these patients. These results and those demonstrating an expression of complement C3 in the proliferative endometrium of patients with endometriosis (5) suggest an intrinsic alteration of the endometrium in this disease. Endometrium of patients with endometriosis is able to synthesize and secrete inflammatory factors and proteins that may alter its ph siological function. These findings will agree also with our previous observations of an increase in a chemotactic factor in the PF of endometriosis patients (3). However, presently no information is Inflammatory changes of the endometrium in endometriosis Fertility and Sterility

5 a; 3: 80 Ill Cl «<.c: c u «< ::E 20 0 glands stroma Figure 4 Chemotactic activity of cellular extracts isolated from luteal biopsies from normal patients. Extracts were prepared from isolated glandular epithelium and stroma and assayed for chemotactic activity using macrophages. Results are presented as mean number of cells per well (±SD). available to indicate whether the PF chemotactic factor and the endometrial chemotactic factor are related. Future studies will need to evaluate the intercycle variations within each patient to be able to generalize further these results. Previous investigators (22) have reported an increase in the number of blood cells infiltrating the endometrium in the late secretory phase as well as in the peri-implantation period. The role of these infiltrating cells in the uterus is not well established (10); their presence in the late secretory phase has suggested a role in early implantation and pregnancy recognition. In fact, one of the mechanisms proposed to explain early pregnancy loss is based on a suppressor immune-cell deficiency or to an increase in macrophage activation and function in the decidua of women with spontaneous abortion (23). Macrophages are routinely present in the connective stroma of several normal tissues. Recently some of its activation factors such as monocyte chemotactic protein-1 (MCP-1), have been shown to be modulated by the endometrial stromal cells, by interleukin (IL-la), tumor necrosis factor (TNF) a or interferon (IFN-a) (Arici A, MacDonald PC, Casey ML, abstract). There is evidence to suggest that MCP-1 is an important determinant of macrophage infiltration into tumors and tissues, and this recruitment is believed to be mediated primarily via chemotaxis. The regulation of MCP-1 and macrophage infiltration by the endometrium may be part of the normal functioning of the tissue by coordinating the recruitment of monocytes only at specific times (24). Other regulatory factors of cellular infiltration include the monocyte-derived colony-stimulating factor (M-CSF) and IL-l. The monocyte-derived colony-stimulating factor is responsible for the promotion of survival, proliferation, and chemotaxis of the cells (24). Interleukin 1 also is associated with chemotaxis, induction of lymphoid infiltration, and activation of inflammatory cells. The expression of IL-l messenger RNA is increased in the secretory phase of the menstrual cycle; however, the exact cellular source of this protein has been difficult to identify due to the large amounts of infiltrating cells (24). Endometrium from patients with endometriosis also exhibits the same inflammatory modifications previously described for the pelvic cavity and the PF. This increase in chemotactic activity throughout the menstrual cycle may be another contributory mechanism for the associated infertility of these patients. Currently, we are investigating the exact nature and characteristics of this chemoattractant. Acknowledgements. We thank Celso-Ramon Garcia, M.D., and Bruce A. Lessey, M.D., for their collaboration with these studies. Complement receptor 3 was provided by John Lambris, Ph.D., Department of Pathology, University of Pennsylvania, Philadelphia, Pennsylvania. REFERENCES 1. Hill JA, Anderson DJ. Lymphocyte activity in the presence of peritoneal fluid from fertile women and infertile women with and without endometriosis. Am J Obstet Gynecol 1989;161: Badawy SZ, Cuenca V, Marshall L, Munchback R, Rinas AC, Coble DA. Cellular components in peritoneal fluid in infertile patients with and without endometriosis. Fertil Steril 1984;42: Leiva MC, Hasty LA, Pfeifer S, Mastroianni L, Lyttle CR. Increased chemotactic activity in peritoneal fluid of patients with mild to moderate endometriosis. Am J Obstet Gynecol 1993; 168: Isaacson KB, Coutifaris C, Garcia CR, Lyttle CR. Production and secretion of complement component 3 by endometriotic tissue. J Clin Endocrinol Metab 1989;69: Isaacson KB, Galman M, Coutifaris C, Lyttle CR. Endometrial synthesis and secretion of complement component-3 by patients with and without endometriosis. Fertil Steril 1990;53: Hasty LA, Lambris JD, Lessey BA, Pruksananonda K, Lyttle CR. Hormonal regulation of complement components and its receptors throughout the menstrual cycle. Am J Dbstet and Gynecol. 1994;170: Lessey BA, Metzger DA, Haney AF, McCarty KS. Immuno- Vol. 62, No.5, November 1994 Leiva et al. Inflammatory changes of the endometrium in endometriosis 971

6 histochemical analysis of estrogen and progesterone receptors in endometriosis: comparison with normal endometrium during the menstrual cycle and the effect of medical therapy. Fertil Steril1989;51: Kamat BR, Isaacson PG. The immunocytochemical distribution of leukocytic subpopulations in human endometrium. Am J Pathol1987;127: Chouddhuri R, Wood GW. Determination of the number and distribution of macrophages, lymphocytes and granulocytes in the mouse uterus after mating through implantation. J Leukocyte Bioi 1991;50: Croy BA, Chapeau C, Reed N, Stewart IJ, PeelS. Is there an essential requirement for bone-marrow-derived cells at the fetomaternal interface during successful pregnancy? A study of pregnancies in immunodeficient mice. In: Wegmann TG, Gill TJ, Nisbet-BrownE, editors. Molecular and cellular immunobiology of the maternal fetal interface. London: Oxford University Press, 1991: Olive DI, Haney AF, Weinberg JB. The nature of the intraperitoneal exudate associated with infertility: peritoneal fluid and serum lysozyme activity. Fertil Steril 1987;48: Halme J, White C, Kauma S, Estes J, Haskill S. Peritoneal macrophages from patients with endometriosis release growth factor activity in vitro. J Clin Endocrinol Metab 1988;66: Hill JA, Faris HM, Schiff I, Anderson DJ. Characterization of leukocyte subpopulations in the peritoneal fluid of women with endometriosis. Fertil Steril 1988;50: Steinleitner A, LambertH, Kazensky C, Danks P. Peritoneal fluid from endometriosis patients affects reproductive outcome in an in vivo model. Fertil Steril 1990;53: The American Fertility Society. Revised American Fertility Society classification of endometriosis. Fertil Steril 1985;43: Leiva MC, Lyttle CR, Jellinck PH. Complement C3 synthesis, peroxidase activity and eosinophil chemotaxis in the rat uterus: effect of estradiol and testosterone. Mol Cell Endocrinol1991;81: Howe RS, Fischkoff SA, Rossi RM, Lyttle CR. Chemotactic capabilities of HL-60 human myeloid leukemia cells differentiated to eosinophils. Exp Hematol 1990;18: Sundstromm C, Nilsson K. Establishment and characterization of a human histiocytic lymphoma cell line (U937). Int J Cancer 1976;17: Kay GE, Lane BC, Snyderman R. Induction of selective biological responses to chemoattractants in human monocyte-like cell line. Infect Immunol1983;41: Satyaswaroop PG, Bressler B, de Ia Penna MM, Gurpide E. Isolation and culture of human endometrial glands. J Clin Endocrinol Metab 1979;48: Lee YH, Howe RS, Sha S, Teuscher C, Lyttle CR. Estrogen regulation of an eosinophil chemotactic factor in the rat uterus. Endocrinology 1990;125; Bulmer JN, Morrison L, Longfellow M, Ritson A. Granulated lymphocytes in human endometrium: further histochemical and immunohistochemical studies. Hum Reprod 1991;6: Hill JA. Immunologic recurrent abortion. In: Recurrent pregnancy loss. Infertility and Reproductive Medicine Clinics of North America. 1991;2: Tabibzadeh S. Human endometrium: an active site of cytokine production and action. Endocr Rev 1991;12: Leiva et al. Inflammatory changes of the endometrium in endometriosis Fertility and Sterility